Electronics > Beginners
TL071 distortion - bad amplifier design?
<< < (24/26) > >>
Kleinstein:
There is not very much more to say to the SSBA circuit, after the corrected annotations. A simple current limitation for protection might be a good idea and could be done with little effort in the standard way.

For the OP based circuit , the simple local feedback version shown by Hero999 could be a good idea for less cross over distortion.

To reduce the problem that the current in the extra voltage amplification stage depends on the supply voltage, one could add diodes in series to R7/ R8 and make those resistors a little smaller. The transistor / diode combination would than be simplified current mirrors.
There would be still some supply dependence, but less (maybe half or a third).

The more radical way would be to have one side a fixed current source and only use the OP to drive one side. Power supply suppression would still be a problem. A transistor in base configuration could help to transfer the signal as a current from ground references (at the OP) to supply referenced for Q5 or Q6.

Zero999:

--- Quote from: spec on December 17, 2018, 05:01:35 am ---
--- Quote from: Hero999 on December 16, 2018, 06:56:41 pm ---Constant current generator #2 should be current mirror.
--- End quote ---
That would be like calling the the input stage a long tailed pair- not very helpful or informative.
That arrangement of Q10 and Q11 has two functions, which if possible, will be explained. Just a clue: a current mirror is a constant current generator.

--- Quote from: Hero999 on December 16, 2018, 06:56:41 pm ---A voltage to current converter would be a transconductance amplifier i.e. voltage in, current proportional to the input voltage out.  It's really a unity gain buffer or current booster circuit i.e. input voltage = output voltage, with a high input impedance and low output impedance.
--- End quote ---
Incorrect, the output transistor pairs convert an input voltage from the VAS to a current to drive the speaker (V/R).
It is difficult to chose titles for circuit functions, but Current Booster, gives totally the wrong message.
'unity gain current booster' that is a contradiction in terms.

--- End quote ---

* The input stage is a long tailed pair.
* Yes a current mirror can be used as a constant current source/sink, but that's not its function here, because the current isn't constant, but proportional to Q3's collector current.

* This circuit uses voltage drive for the speaker, not current drive, hence the output stage provides current gain, not voltage to current conversion.
If I'm wrong and you're right, then why does entering in the terms I've given into a Google image search yield circuit elements, similar to those detailed in the Self Simple Blameless Amplifier schematic, yet the ones you've listed give something completely different?

Current mirror - the first hit looks remarkably how Q10 and Q11 are connected.

Constant current generator - mostly different results. A couple match, but nowhere near as much as looking for the correct term: current mirror.

Voltage to current converter - looks nothing like the output stage. Clicking on the links also reveals descriptions of circuits which generate a current, proportional to an input voltage, with a few the other way round: very confusing!

I admit, looking for current booster unity gain buffer doesn't give exactly the same circuit as the output stage, but with the exception of a couple rouge hits, it's fairly close (emitter followers). I admit I could have chosen a better term: push-pull current booster, push-pull unity gain buffer or class AB output stage give better hits, although the last one isn't quite as descriptive on its own.

I don't want to turn this into a debate about semantics, but using vaguely the correct terms makes it easier for newcomers to search for things and avoids confusion. I admit it can be difficult to choose descriptive terms for such abstract concepts.


--- Quote from: Kleinstein on December 17, 2018, 08:24:42 am ---There is not very much more to say to the SSBA circuit, after the corrected annotations. A simple current limitation for protection might be a good idea and could be done with little effort in the standard way.

For the OP based circuit , the simple local feedback version shown by Hero999 could be a good idea for less cross over distortion.

To reduce the problem that the current in the extra voltage amplification stage depends on the supply voltage, one could add diodes in series to R7/ R8 and make those resistors a little smaller. The transistor / diode combination would than be simplified current mirrors.
There would be still some supply dependence, but less (maybe half or a third).
--- End quote ---

Yes, that seems sensible.


--- Quote ---The more radical way would be to have one side a fixed current source and only use the OP to drive one side. Power supply suppression would still be a problem. A transistor in base configuration could help to transfer the signal as a current from ground references (at the OP) to supply referenced for Q5 or Q6.

--- End quote ---
Similar to what speck proposed earlier on in the thread?
https://www.eevblog.com/forum/beginners/tl071-distortion-bad-amplifier-design/msg2030107/#msg2030107
Kleinstein:

--- Quote from: Hero999 on December 17, 2018, 10:44:41 am ---........

--- Quote ---The more radical way would be to have one side a fixed current source and only use the OP to drive one side. Power supply suppression would still be a problem. A transistor in base configuration could help to transfer the signal as a current from ground references (at the OP) to supply referenced for Q5 or Q6.

--- End quote ---
Similar to what speck proposed earlier on in the thread?
https://www.eevblog.com/forum/beginners/tl071-distortion-bad-amplifier-design/msg2030107/#msg2030107

--- End quote ---
Yes exactly that kind of circuit I was thinking of. Somehow forgot it was already in this thread.
not1xor1:

--- Quote from: Hero999 on December 17, 2018, 10:44:41 am ---[...]
Similar to what speck proposed earlier on in the thread?
https://www.eevblog.com/forum/beginners/tl071-distortion-bad-amplifier-design/msg2030107/#msg2030107

--- End quote ---

who is speck ?  :scared:
According to wikipedia you are probably hungry...  ;D
His nickname is spec, no K
Zero999:

--- Quote from: not1xor1 on December 17, 2018, 11:22:19 am ---
--- Quote from: Hero999 on December 17, 2018, 10:44:41 am ---[...]
Similar to what speck proposed earlier on in the thread?
https://www.eevblog.com/forum/beginners/tl071-distortion-bad-amplifier-design/msg2030107/#msg2030107

--- End quote ---

who is speck ?  :scared:
According to wikipedia you are probably hungry...  ;D
His nickname is spec, no K

--- End quote ---
Sorry spec. I don't know why I added the k.:palm:

Yes, I've tried that stuff when I was in Venice. It's a bit on the salty side for my taste, but a a little on a pizza is OK.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod